0.28.0 — operational hardening (plan D1–D5): CI deploy gate, versioned images, rate limiting, resource limits, health probes
Some checks failed
ci / build-test (push) Has been cancelled
ci / docker-build (push) Has been cancelled
ci / web (push) Has been cancelled
Desktop App / Build Linux (amd64) (push) Has been cancelled
Desktop App / Attach to Release (push) Has been cancelled

D1: deploy.sh CI gate — read-only SHA via git ls-remote, Gitea commit-status
    poll, portable mkdir deploy lock (macOS, no flock), TOCTOU guard, token
    passed via curl --config - (not argv), graceful misconfig tolerance.
D2: version-tagged images — OIKOS_VERSION=v$VERSION, keep-last-3 prune derived
    from 'docker compose config --images'; VERSION read after pull.
D3: per-IP rate limiting — new internal/httpapi/ratelimit.go (x/time/rate),
    rightmost-XFF, /healthz exempt, ctx-driven sweep; disabled by default.
D4: mem_limit/cpus on all 10 compose services.
D5: staleness-aware health probes — new internal/health package wired into
    scheduler (:8093) and notifier (:8094); nomos already had :8092.

Two /review passes hardened the deploy lock, TOCTOU guard, token hygiene,
and XFF handling.
This commit is contained in:
2026-08-08 21:31:16 +02:00
parent ef762794e7
commit fa79c1ea25
14 changed files with 852 additions and 41 deletions

View File

@@ -18,6 +18,7 @@ import (
"github.com/dtoro/oikos/internal/config"
"github.com/dtoro/oikos/internal/db"
"github.com/dtoro/oikos/internal/health"
"github.com/google/uuid"
)
@@ -25,7 +26,14 @@ import (
func Run(ctx context.Context, pool *db.Pool, cfg config.Config) {
slog.Info("notifier: starting")
// Liveness probe (plan D5): the notifier ticks every 15s (approvals) and
// 30s (reactions). 2 min staleness covers a slow Matrix round-trip plus a
// missed tick without false-failing.
probe := health.New(2 * time.Minute)
probe.Serve(ctx, cfg.HealthListen)
processPendingApprovals(ctx, pool, cfg)
probe.Bump()
ticker := time.NewTicker(15 * time.Second)
defer ticker.Stop()
@@ -40,8 +48,10 @@ func Run(ctx context.Context, pool *db.Pool, cfg config.Config) {
return
case <-ticker.C:
processPendingApprovals(ctx, pool, cfg)
probe.Bump()
case <-reactionTimer.C:
pollReactions(ctx, pool, cfg)
probe.Bump()
}
}
}
@@ -52,13 +62,13 @@ func RunnerForMain() func(context.Context, *db.Pool, config.Config) {
}
type pendingApproval struct {
ID uuid.UUID
Action string
RiskClass string
TokenHash *string
AlertSentAt *time.Time
MatrixEventID *string
ExpiresAt time.Time
ID uuid.UUID
Action string
RiskClass string
TokenHash *string
AlertSentAt *time.Time
MatrixEventID *string
ExpiresAt time.Time
}
// processPendingApprovals finds pending approvals, generates tokens, and sends Matrix alerts.
@@ -173,7 +183,7 @@ func checkReaction(ctx context.Context, cfg config.Config, roomID, eventID strin
var result struct {
Chunk []struct {
Type string `json:"type"`
Type string `json:"type"`
Content struct {
RelatesTo map[string]string `json:"m.relates_to"`
} `json:"content"`
@@ -259,7 +269,9 @@ func sendMatrixAlert(ctx context.Context, cfg config.Config, approvalID uuid.UUI
}
defer resp.Body.Close()
var mxResp struct{ EventID string `json:"event_id"` }
var mxResp struct {
EventID string `json:"event_id"`
}
json.NewDecoder(resp.Body).Decode(&mxResp)
if mxResp.EventID == "" {